Systems and methods for clustering of customers using transaction patterns

Patent metadata

Field	Value
Publication number	US-10685341-B2
Application number	US-201916544876-A
Country	US
Kind code	B2
Filing date	Aug 19, 2019
Priority date	May 30, 2013
Publication date	Jun 16, 2020
Grant date	Jun 16, 2020

Abstract

A system can include one or more non-transitory computer-readable media storing computing instructions configured to run on one or more processors and perform certain acts. The acts can include receiving, at a back-end device of the system, a query from a user interface of a front-end device requesting two or more user identifications mapped to a same payment option of multiple payment options. The acts also can include distilling a subset of user identifications from among the multiple user identifications. Additionally, the acts can include identifying a first classification comprising a single user of the subset of user identifications. Further, the acts can include identifying a second classification comprising two or more first users of the subset of user identifications that are part of a same household. Other embodiments are disclosed.

First claim

What is claimed is:

1. A system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computer instructions configured to run on the one or more processors and to perform:
receiving, at a back-end device of the system, a query from a user interface of a front-end device requesting two or more user identifications mapped to a same payment option of multiple payment options;
in association with multiple transactions, receiving multiple user identifications of multiple users and their respective payment options of the multiple payment options;
distilling a subset of payment options from among the multiple payment options;
distilling a subset of user identifications from among the multiple user identifications, such that each respective payment option of the subset of payment options is associated with more than one user identification of the subset of user identifications;
analyzing the subset of payment options to create an analysis of patterns based on the subset of user identifications associated with a single payment option of the subset of payment options;
identifying, using the analysis, a first classification comprising a single user of the subset of user identifications that is mapped to more than one user identification of the subset of user identifications;
identifying, using the analysis, a second classification comprising two or more first users of the subset of user identifications that are part of a same household; and
responsive to receiving the query, sending instructions to display, on the user interface of the front-end device, at least one of the first classification or the second classification for each respective user identification of the subset of user identifications.

2. The system of claim 1 , wherein the computer instructions are further configured to perform:
identifying the multiple user identifications associated with multiple user profiles.

3. The system of claim 1 , wherein identifying, using the analysis, the first classification further comprises:
determining, using a logistic regression algorithm, the first classification.

4. The system of claim 3 , wherein identifying, using the analysis, the second classification further comprises:
determining, using the logistic regression algorithm, the second classification.

5. The system of claim 3 , wherein the determining, using the logistic regression algorithm, further comprises filtering the subset of user identifications into the first classification or the second classification using probabilistic matching by the logistic regression algorithm expressed as:

P
⁡

(

I
t
ij

=

1
❘

c
t
i

,

c
t
j

)

=

1

1
+

e

-
wx

where:
c i t represents a i th user identification in a record D t ,
c j t represents a j th user identification in the record D t ,
P(i t ij c t i , c t j ) is a posterior probability to be calculated,
I t ij represents a binary random variable that denotes whether two users c i t and c j t in the record D t belong to the first classification (I t ij =1) or otherwise,
x represents a vector representing features derived from user identifications in c i t and c j t , and
w represents weights of a logistic regression.

6. The system of claim 1 , wherein the computer instructions are further configured to perform:
filtering the each respective user identification of the subset of user identifications comprising:
determining whether a record D t is classified in the first classification based on analyzing the each respective payment option of the subset of payment options by determining a respective first similarity score for each of the multiple user identifications of the subset of user identifications mapped to a same payment option, using:
let D t ={D i }, i=1, 2, . . . N, where in D i denotes a collection of information pertaining to the multiple users who used a particular payment option t i , and N denotes a number of records in a data set; and
each respective record D k may be expressed as {t k , c 1 k c 2 k , . . . }, where D k includes a t k unique user identification generated from the particular payment option t i comprising a credit card, and c j k represents a jth user identification that used the t k unique user identification; and

 determining whether the record D t is classified in the second classification by using the one or more processors to determine similarities in user information and filtering for similarities using a second similarity score, wherein the determining whether the record D t is classified in the second classification further comprises:
searching the user information comprising at least first names, last names, and physical addresses;
applying an edit distance on the first names and the last names between all pairs of the user information in the record D t that shared the single payment option;
filtering the user information by the single payment option attributable to the multiple user identifications and by transaction history; and
clustering the user information into the second classification when the edit distance for the user information falls below a preset threshold.

7. The system of claim 6 , wherein the respective first similarity score for the each of the multiple user identifications of the subset of user identifications is indicative of similarities between a string of one or more pieces of the user information associated with a respective one of the multiple user identifications and a respective string of the one or more pieces of the user information associated with each of one or more other ones of the multiple user identifications.

8. The system of claim 7 , wherein the one or more pieces of the user information comprises an email address.

9. The system of claim 1 , wherein the computer instructions are further configured to perform:
filtering the each respective user identification of the subset of user identifications comprising:
filtering a first user identification of the subset of user identifications as one of the multiple user identifications associated with a first u

Assignees

• Walmart Apollo Llc

Inventors

• Ali Afroza
• Achan Kannan
• Benson Ron

Classifications

• G06Q20/22 Primary

  Payment schemes or models · CPC title

• G06Q20/227

  characterised in that multiple accounts are available, e.g. to the payer · CPC title

• G06Q40/00

  Finance; Insurance; Tax strategies; Processing of corporate or income taxes · CPC title